create-spec
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect instructions through external data ingestion.
- Ingestion points: Fetches content from external GitHub issues using the
gh issue viewcommand. - Boundary markers: No explicit markers or instructions are provided to the agent to distinguish between the issue's data and the skill's operational instructions.
- Capability inventory: The skill possesses the
Writetool andBashtool, allowing for file modification and command execution based on processed data. - Sanitization: There is no evidence of filtering or sanitizing the input retrieved from GitHub before it is used to determine the specification logic.
- [COMMAND_EXECUTION]: Utilizes the
ghCLI tool to interact with GitHub services. - Evidence: Step 1 explicitly instructs the use of
gh issue viewto retrieve context for the feature specification. - [EXTERNAL_DOWNLOADS]: Accesses external resources to retrieve feature requirements.
- Evidence: Downloads data from GitHub via the CLI when an issue number or URL is provided by the user.
Audit Metadata